Personalized training scheme generation system and method based on data analysis

Abstract

The invention discloses a personalized training scheme generation system and method based on data analysis, and relates to the technical field of training management. The personalized training scheme generation system based on data analysis comprises a training portrait initial evaluation module, a portrait reconstruction judgment module, a training path timeliness judgment module and a reconstruction path validity evaluation module. According to the method, the initial portrait updating priority is obtained, whether portrait updating is triggered or not is judged for employee training portraits, if yes, portrait construction data is updated, if not, updating critical evaluation is carried out, then whether personalized training paths are reconstructed or not is judged, and if not, personalized training path generation intervals are optimized, so that the personalized training path generation efficiency is improved. Otherwise, whether the initial portrait updating priority is adjusted or not is judged, the generation timeliness of the personalized training scheme is improved, and the problem that the association degree between the generation timeliness of the personalized training scheme and the training progress is insufficient in the prior art is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent training management, and particularly relates to a personalized training plan generation system and method based on data analysis. BACKGROUND

[0002] With the rapid development of data analysis technology and the widespread application of artificial intelligence, personalized training plans have gradually become an important means to improve employee skills and efficiency. Based on data analysis, personalized training plan generation systems use data mining, machine learning, and other technologies to tailor training content and optimize learning paths based on the characteristics and learning progress of learners. This approach can achieve precise and real-time allocation of learning resources, helping learners maximize their learning effectiveness while saving training costs for enterprises and improving overall talent quality.

[0003] Existing personalized training plan generation technologies mainly rely on learners' basic information and past learning records to build recommendation systems. Common methods include content-based recommendation, collaborative filtering, and hybrid recommendation algorithms. However, these methods focus on static data analysis and ignore dynamic changes and individual differences in the learning process. Although some systems incorporate machine learning and data mining techniques, they still face issues such as insufficient data accuracy, real-time performance, and personalized depth, making it difficult to fully meet the needs of individual differences and efficient learning path generation.

[0004] For example, the patent application with publication number CN118761874A discloses a personalized training plan generation method based on big data analysis, which includes learning behavior analysis, personalized content generation, dynamic feedback adjustment, and adaptive path optimization. The results of learning behavior analysis serve as the data basis for personalized content generation, and dynamic feedback adjustment signals serve as data support for adaptive path optimization. Adaptive path optimization involves using reinforcement learning algorithms to adaptively adjust learning paths based on learning data and dynamic feedback information, optimizing learning order and content arrangement.

[0005] For example, the invention patent announcement No. CN118396804B, a precise teaching management method and system based on adaptive learning analysis, includes: for each student, determine the learning track data and obtain the learning resource browsing record, combine to get multi-source heterogeneous learning data, determine the knowledge node and the cognitive state node and build the personalized knowledge graph; determine the optimal learning path, combine multi-task learning for joint optimization, apply multi-view attention network, determine the planning path influence factor and recommend learning resources, generate the initial learning behavior sequence, update the personalized knowledge graph in real time, until the preset maximum iteration number is reached, get the learning behavior sequence; based on the learning behavior sequence, deep self-encoding is performed, the student learning portrait is constructed, the student learning effect is diagnosed and analyzed in multiple dimensions, the individualized diagnosis report is generated, the adaptive evaluation model is constructed, the corresponding diagnostic evaluation questions are generated, and the individualized teaching scheme is generated.

[0006] However, in the process of implementing the technical scheme of the present application, the applicant found that the above-mentioned technology at least has the following technical problems:

[0007] In the prior art, since the employee training portrait is the core part of the personalized training path, the update of the related data involved in the construction of the employee training portrait is not fine enough, which reduces the timeliness of the constructed employee training portrait, and further reduces the timeliness of the personalized training path, and there is a problem that the timeliness of the generation of the personalized training scheme is not sufficient in association with the training progress. SUMMARY

[0008] The embodiments of the present application provide a personalized training scheme generation system and method based on data analysis, which solves the problem of insufficient association between the timeliness of the generation of the personalized training scheme and the training progress in the prior art, and improves the timeliness of the generation of the personalized training scheme.

[0009] The embodiment of the application provides a personalized training scheme generation system based on data analysis, which comprises a training portrait initial evaluation module, a portrait reconstruction judgment module, a training path timeliness judgment module and a reconstruction path effectiveness evaluation module; wherein the training portrait initial evaluation module is used for obtaining initial portrait update priorities of various portrait construction data from a preset database, and performing training progress change evaluation on an employee training portrait in combination with change conditions of the various portrait construction data; the employee training portrait is used for measuring progress change of the employee in an actual training process; the portrait reconstruction judgment module is used for judging whether to trigger portrait update based on the obtained evaluation result; if yes, the portrait construction data of the corresponding category is updated, otherwise, update critical evaluation is performed on the various portrait construction data; the training path timeliness judgment module is used for judging whether to reconstruct the employee training portrait based on the obtained evaluation result, constructing a personalized training path based on the obtained employee training portrait, and performing update timeliness quantization on the employee training portrait; the reconstruction path effectiveness evaluation module is used for judging whether to reconstruct the personalized training path based on the quantization result; if no, the personalized training path generation interval is optimized, otherwise, reconstruction effectiveness evaluation is performed on the reconstructed personalized training path to judge whether to adjust the initial portrait update priority.

[0010] In the embodiment of the application, a personalized training scheme generation method based on data analysis is provided, and the specific steps are as follows: obtaining initial portrait update priorities of various portrait construction data from a preset database, and performing training progress change evaluation on an employee training portrait in combination with change conditions of the various portrait construction data; the employee training portrait is used for measuring progress change of the employee in an actual training process; judging whether to trigger portrait update based on the obtained evaluation result; if yes, the portrait construction data of the corresponding category is updated, otherwise, update critical evaluation is performed on the various portrait construction data; judging whether to reconstruct the employee training portrait based on the obtained evaluation result, constructing a personalized training path based on the obtained employee training portrait, and performing update timeliness quantization on the employee training portrait; judging whether to reconstruct the personalized training path based on the quantization result; if no, the personalized training path generation interval is optimized, otherwise, reconstruction effectiveness evaluation is performed on the reconstructed personalized training path to judge whether to adjust the initial portrait update priority.

[0011] The one or more technical solutions provided in the embodiment of the application have at least the following technical effects or advantages:

[0012] 1、Through the initial image update priority obtained, the training progress change of the employee training image is evaluated by combining the changes of various image construction data, and it is judged whether to trigger image update, if yes, update, otherwise, update critical assessment, then judge whether to reconstruct the employee training image based on the evaluation result, then reconstruct the personalized training path, and update the time limit for the employee training image, and judge whether to reconstruct the personalized training path, if not, optimize the personalized training path generation interval, otherwise, judge whether to adjust the initial image update priority, thereby improving the timeliness of the employee training image, and further improving the timeliness of the personalized training scheme, effectively solving the problem that the timeliness and training progress correlation of the personalized training scheme in the prior art are insufficient.

[0013] 2、Through the image construction data change parameter obtained, and the image construction change allocation amount obtained from the preset database, the image construction data change parameter and the image construction change allocation amount are coupled to obtain the employee training progress change influence amount after weighting operation, thereby more accurately quantifying the influence degree of the employee training index on the employee training image, and further improving the timeliness of the employee training image.

[0014] 3、Through the training path progress data obtained, and the training path progress allocation rate and the reference path progress time data obtained from the preset database, then the average training index completion time and the reference training index completion time are subjected to ratio operation to obtain the training index completion time value, and the training participation time and the reference training participation time are subjected to ratio operation to obtain the training participation value, and the training path effectiveness evaluation index is obtained based on the above data, thereby more accurately quantifying the effectiveness degree of the optimized and reconstructed personalized training path, and further improving the timeliness of the personalized training path. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 The structure schematic diagram of the personalized training scheme generation system based on data analysis provided by the embodiment of the present application;

[0016] Figure 2 The flowchart of the update critical assessment of various image construction data provided by the embodiment of the present application;

[0017] Figure 3 The flowchart of judging whether to adjust the initial image update priority provided by the embodiment of the present application;

[0018] Figure 4 The flowchart of the personalized training scheme generation method based on data analysis provided by the embodiment of the present application;

[0019] Figure 5An interface diagram of a personalized training plan generation system based on data analysis provided by an embodiment of the present application;

[0020] Figure 6 An interface diagram of a training portrait initial evaluation module of a personalized training plan generation system based on data analysis provided by an embodiment of the present application;

[0021] Figure 7 An interface diagram of a portrait reconstruction determination module of a personalized training plan generation system based on data analysis provided by an embodiment of the present application;

[0022] Figure 8 An interface diagram of a training path timeliness determination module of a personalized training plan generation system based on data analysis provided by an embodiment of the present application;

[0023] Figure 9 An interface diagram of a reconstruction path effectiveness evaluation module of a personalized training plan generation system based on data analysis provided by an embodiment of the present application. DETAILED DESCRIPTION

[0024] The embodiment of the present application provides a personalized training plan generation system and method based on data analysis, solves the problem of insufficient correlation between the timeliness of the generation of the personalized training plan and the training progress in the prior art, obtains an initial portrait update priority of various portrait construction data, classifies employee portrait construction data, obtains a historical average update frequency corresponding to the various portrait construction data, maps the initial portrait update priority of the corresponding category portrait construction data in a preset database based on the historical average update frequency, then monitors the change of the various portrait construction data to obtain a corresponding employee training progress change influence quantity, then determines the employee training progress change influence quantity and a progress change critical value obtained from the preset database, if the employee training progress change influence quantity is not less than the progress change critical value, the portrait construction data of the corresponding category is updated, otherwise, the change of the portrait construction data of the corresponding category is continuously monitored, then whether the portrait update is triggered is determined based on the obtained evaluation result, if yes, the portrait construction data is updated, otherwise, an update critical assessment is performed on the various portrait construction data, finally, whether the employee training portrait is reconstructed is determined based on the obtained evaluation result, a personalized training path is constructed based on the obtained employee training portrait, and the update timeliness of the employee training portrait is quantified, and whether the personalized training path is reconstructed is determined based on the result, if not, the personalized training path generation interval is optimized, otherwise, whether the initial portrait update priority is adjusted is determined, and the timeliness of the generation of the personalized training plan is improved.

[0025] The technical solution in the embodiment of the present application solves the problem of insufficient correlation between the timeliness of the generation of the personalized training plan and the training progress, and the general idea is as follows:

[0026] The initial image update priority of various image construction data is obtained, the training progress change of the employee training image is evaluated to determine whether to trigger image update, if yes, the image construction data is updated, otherwise, the update criticality is evaluated, then it is determined whether to rebuild to build a personalized training path, and the update timeliness quantification is performed to determine whether to rebuild the personalized training path, if no, the personalized training path generation interval is optimized, otherwise, it is determined whether to adjust the initial image update priority, thereby improving the timeliness of generating the personalized training scheme.

[0027] In order to better understand the above technical solutions, the above technical solutions will be described in detail in the following in combination with the drawings of the specification and specific embodiments.

[0028] As shown in Figure 1 , it is a structure schematic diagram of the personalized training scheme generation system based on data analysis provided by the embodiment of the application, the personalized training scheme generation system based on data analysis provided by the embodiment of the application includes: a training image initial evaluation module, an image reconstruction determination module, a training path timeliness determination module and a reconstruction path effectiveness evaluation module.

[0029] The training image initial evaluation module is used to obtain the initial image update priority of various image construction data from a preset database, and to evaluate the training progress change of the employee training image in combination with the change of various image construction data, the employee training image is used to measure the progress change of the employee in the actual training process, and the initial image update priority is used to initially quantify the update demand degree of various image construction data.

[0030] The image reconstruction determination module is used to determine whether to trigger image update based on the obtained evaluation result, if yes, the image construction data of the corresponding category is updated, otherwise, the update criticality of various image construction data is evaluated, the update criticality is used to determine whether all image construction data meets the requirements of critical update.

[0031] The training path timeliness determination module is used to determine whether to rebuild the employee training image based on the obtained evaluation result, to build a personalized training path based on the obtained employee training image, and to quantize the update timeliness of the employee training image, the personalized training path represents the planning route of the employee training, and the update timeliness quantification is used to quantify the timeliness requirement compliance degree of the current employee training image and the personalized training path.

[0032] The reconstruction path validity evaluation module is configured to determine whether to reconstruct the personalized training path based on the quantization result. If not, the personalized training path generation interval is optimized. Otherwise, the reconstruction effectiveness of the reconstructed personalized training path is evaluated to determine whether to adjust the initial image update priority. The reconstructed personalized training path represents a personalized training path reconstructed based on the initial personalized training path. The reconstruction effectiveness evaluation is configured to evaluate the improvement degree of the reconstructed personalized training path on the employee training progress.

[0033] In the embodiment, when the update of the related data involved in the construction of the employee training image by the data analysis based personalized training scheme generation system is not fine, the timeliness of the constructed employee training image and the personalized training path may be reduced, and the timeliness of the generation of the personalized training scheme and the correlation degree with the training progress are insufficient. Through the closed-loop analysis of the feedback of the employee training image and the personalized training path, the timeliness of the employee training image is improved, thereby improving the timeliness of the personalized training path, and further improving the generation timeliness of the personalized training scheme.

[0034] It should be noted that before designing the data analysis based personalized training scheme generation system, a preset database for storing various types of setting data is established by preset personnel. The preset database includes but is not limited to initial image update priority, progress change threshold, image construction change allocation, preset threshold, path timeliness influence limit, path timeliness allocation value, reference image update time, path generation interval influence multiple, path-image compliance judgment value, training path progress allocation rate and reference path progress time data, etc. Various numerical values are directly set by preset professionals. For example, the path-image compliance judgment value is obtained by substituting the training path progress data corresponding to the case where the personalized training path exists reconstruction in the historical database into the specific limit expression of the training path validity evaluation index to obtain the corresponding data set. The path-image compliance judgment value is obtained by mean operation on the data set, and is stored in the preset database in advance.

[0035] As a further scheme, the specific steps of training progress change evaluation of the employee training image are as follows:

[0036] Step one, classify the employee image construction data, and obtain the historical average update frequency corresponding to each type of image construction data. Based on the historical average update frequency, the initial image update priority of the corresponding type of image construction data is mapped in the preset database.

[0037] It should be added that the priority mapping set of the historical average update frequency of the image construction data of each category and the initial image update priority is inputted with the category corresponding to the image construction data, and the initial image update priority corresponding to the image construction data is outputted. The priority mapping set represents a set of mapping relationships between the historical average update frequency of the image construction data of each category and the initial image update priority.

[0038] Step two, real-time monitoring of the change of each category of image construction data to obtain the corresponding employee training progress change impact.

[0039] Step three, the employee training progress change impact is judged with the progress change critical value obtained from the preset database: if the employee training progress change impact is not less than the progress change critical value, the image construction data of the corresponding category is updated, otherwise the change of the image construction data of the corresponding category is continuously monitored.

[0040] In this embodiment, by real-time monitoring of the change of each category of image construction data, the system can timely capture the progress change of employee training, and through the dynamic monitoring, the lag of employee training progress can be effectively avoided, which is helpful for timely adjusting the training scheme; the method updates the image according to the training progress change of the employee, so that the image update of each employee is more personalized.

[0041] As a further scheme, the specific acquisition method of the employee training progress change impact is as follows:

[0042] Firstly, the image construction data change parameter is obtained, and the image construction data change parameter includes the initial image update priority and the training index change number.

[0043] It should be added that the initial image update priority is obtained from the preset database, and the training index change number is obtained by counting the number of training indexes of the employee that need to be trained through the system log.

[0044] Then, the image construction change allocation amount is obtained from the preset database, and the image construction change allocation amount includes the initial priority allocation amount and the training index allocation amount.

[0045] Specifically, the image construction change allocation quantity is obtained from a preset database, and the image construction change allocation quantity represents an influence degree of an image construction data change parameter on an influence amount of employee training progress change. Each image construction data change parameter and the image construction change allocation quantity have a unique mapping relationship, and a value range is between 0 and 1; for example, a mapping set of the image construction data change parameter and the preset image construction change allocation quantity is obtained by inputting the real-time initial image update priority and the training index change quantity into the mapping set to obtain a corresponding initial priority allocation quantity and a training index allocation quantity, which represent an influence degree of the initial image update priority and the training index change quantity on the influence amount of employee training progress change, and a sum of the two is 1.

[0046] Finally, the employee training progress change influence amount is obtained by performing weighting operation and coupling on the image construction data change parameter and the image construction change allocation quantity, and the employee training progress change influence amount is used to quantify an influence degree of the employee training index on the employee training image.

[0047] A specific limit expression of the employee training progress change influence amount is as follows:

[0048]

[0049] In the formula, IPU represents an initial image update priority of the image construction data of the corresponding category, CTI represents a training index change quantity of the image construction data of the corresponding category, μ represents the initial priority allocation quantity, represents the training index allocation quantity, and ETP represents the employee training progress change influence amount of the image construction data of the corresponding category.

[0050] In the embodiment, the algorithm analyzes the image construction data change parameter and the image construction change allocation quantity to obtain the employee training progress change influence amount. In the formula, when the initial image update priority is greater, it indicates that the update demand of the image construction data of the corresponding category is greater, the update influence degree on the employee training image is higher, and the employee training progress change influence amount is greater. Meanwhile, when the training index change quantity is greater, it indicates that the progress change of the image construction data of the corresponding category is greater, and the employee training progress change influence amount is greater. Through analysis of the employee training progress change influence amount, the influence degree of the employee training index change on the employee training image is quantified more accurately, and the timeliness of generation of the individualized training scheme is improved.

[0051] In addition, the image construction data change parameter is not independent, and the initial image update priority and the number of training index changes are related to each other. For example, if the initial image update priority of a certain type of image construction data is higher, the update requirement degree of the image construction data may be higher, the update frequency in the historical data may be higher, and the change of the image construction data may be more, and the number of training index changes of the image construction data may be more.

[0052] As shown in Figure 2 The flowchart of the critical evaluation of the update of various types of image construction data provided by the embodiment of the application is shown in the figure. The specific logic is as follows: first, the progress change difference of various types of image construction data that have not been updated is obtained, and each progress change difference is compared with a preset critical value; then, if the progress change difference of the corresponding type of image construction data is lower than the preset critical value, the progress update value of the corresponding type of image construction data is recorded as a first progress update value, otherwise the progress update value of the corresponding type of image construction data is recorded as a second progress update value; finally, the progress update values of all types of image construction data are counted to obtain a progress update cumulative value, and if the progress update cumulative value is higher than a preset cumulative value, the employee training image is updated, otherwise no additional processing is performed; through the above process, the timeliness of the employee training image obtained by quantifying the determination of the employee training progress change influence quantity is more accurately quantified, so that the corresponding update optimization is timely taken, and the timeliness of the generation of the individualized training scheme is improved.

[0053] As a further scheme, the specific process of the critical evaluation of the update of various types of image construction data is as follows:

[0054] H1, the progress change difference of various types of image construction data that have not been updated is obtained, and each progress change difference is compared with a preset critical value obtained from a preset database. The progress change difference represents the difference between the employee training progress change influence quantity and the progress change critical value.

[0055] Specifically, the preset critical value is preset by a preset staff and stored in the preset database.

[0056] H2, if the progress change difference of the corresponding type of image construction data is lower than the preset critical value, the progress update value of the corresponding type of image construction data is recorded as a first progress update value, otherwise the progress update value of the corresponding type of image construction data is recorded as a second progress update value.

[0057] The progress update value represents a value marked by comparing the progress change difference of each type of portrait construction data with a preset threshold value, which is used to quantify the progress change degree of different comparison conditions; the first progress update value is higher than the second progress update value, for example, the first progress update value is 1 and the second progress update value is 0.

[0058] H3, the progress update cumulative value is obtained by counting the progress update values of all types of portrait construction data, and if the progress update cumulative value is higher than a preset cumulative value, the employee training portrait is updated, otherwise no additional processing is performed.

[0059] In this embodiment, by obtaining the progress change difference of each type of portrait construction data and comparing it with the preset threshold difference, the system can more accurately determine which employee portraits need to be updated in time, improving the timeliness of the employee training portrait. By counting the progress update values of all types of portrait construction data, calculating the progress update cumulative value, and comparing it with the preset cumulative value, it is helpful to further control whether to perform overall update of the employee training portrait. This mechanism ensures that the overall training portrait update will only be performed when the progress update of all employee portraits reaches a certain threshold, avoiding too frequent overall update and ensuring the timeliness of the employee training portrait and improving the timeliness of the generation of the individualized training plan.

[0060] As a further solution, the specific process of updating the employee training portrait to quantify the timeliness is as follows:

[0061] When generating an individualized training path, the update of the current portrait construction data of each type is obtained to obtain an individualized path timeliness influence value of the individualized training path, and the current individualized path timeliness influence value is compared with the path timeliness influence limit value obtained from the preset database.

[0062] Specifically, the path timeliness influence limit value is obtained from the preset database. In one specific embodiment, the preset staff substitutes the training qualified path timeliness evaluation data in the historical database into the specific limit expression of the individualized path timeliness influence value to obtain the corresponding data set, and performs mean operation on the data set to obtain the path timeliness influence limit value.

[0063] On the one hand, if the individualized path timeliness influence value is not greater than the path timeliness influence limit value, the individualized training path generation interval is optimized.

[0064] On the other hand, if the individualized path timeliness influence value is greater than the path timeliness influence limit value, the reconstructed individualized training path is reconstructed and the effectiveness is evaluated.

[0065] In the embodiment, by acquiring the data update of various portraits in real time, the system can dynamically adjust the timeliness of the personalized training path according to the latest portrait of the employee, which ensures that the personalized training path always meets the actual state and needs of the employee training portrait, avoids the fixed time interval training mode, and improves the timeliness of the employee personalized training path.

[0066] As a further solution, the specific acquisition process of the personalized path timeliness influence value is as follows:

[0067] First, acquire the path timeliness evaluation data, which includes the progress update cumulative value, the employee training progress change influence amount, the progress change critical value, and the portrait update time length.

[0068] Specifically, the portrait update time length is obtained through system log records.

[0069] Second, obtain the path timeliness distribution value and the reference portrait update time length from the preset database, the path timeliness distribution value includes the progress update distribution value, the progress determination distribution value, and the portrait update distribution value.

[0070] Specifically, the path timeliness distribution value is obtained from the preset database, and the path timeliness distribution value represents the influence degree of the path timeliness evaluation data on the personalized path timeliness influence value. Each path timeliness evaluation data and the path timeliness distribution value have a unique mapping relationship, and the value range is between 0 and 1; for example, a mapping set of the path timeliness evaluation data and the preset path timeliness distribution value is constructed, and the real-time progress update cumulative value, the employee training progress change influence amount, and the portrait update time length are input into the mapping set to obtain the corresponding progress update distribution value, progress determination distribution value, and portrait update distribution value, which represent the influence degree of the progress update cumulative value, the employee training progress change influence amount, and the portrait update time length on the personalized path timeliness influence value, respectively, and the sum of the three is 1.

[0071] Specifically, the reference portrait update time length is set by the preset staff in advance and stored in the preset database, which represents the maximum limit time length that the portrait update interval time length can accept.

[0072] Third, the employee training progress change influence amount and the progress change critical value are subjected to deviation operation to obtain the progress determination deviation amount.

[0073] It should be explained that the deviation operation means that the employee training progress change influence amount is subtracted from the progress change critical value and then the result is divided by the progress change critical value.

[0074] The specific acquisition method of the progress determination deviation amount is as follows:

[0075]

[0076] Among them, n represents the category number of each type of portrait construction data, n=1,2,...,N, N represents the total number of portrait construction data categories, DP n It represents the impact of employee training progress changes on the nth type of portrait construction data, DP0 represents the critical value of progress change, and DPJ represents the progress judgment deviation. When the impact of employee training progress changes is greater than the corresponding critical value of progress change, the corresponding progress judgment deviation is larger, indicating that the degree of change in employee training progress is higher. At the same time, a larger progress judgment deviation also indicates a higher timeliness of each type of employee training portrait, and is independent of the cumulative value of progress update. The two represent the quantitative values ​​of the timeliness evaluation of employee training portraits under different circumstances. However, when the two are larger, they both indicate that the timeliness of employee training portraits is higher.

[0077] In the fourth step, the progress judgment deviation is weighted and coupled with the cumulative progress update value, the portrait update duration, and the corresponding path timeliness distribution value to obtain the personalized path timeliness impact value. The personalized path timeliness impact value is used to quantify the influence of the current employee training portrait on the personalized training path to be generated.

[0078] The specific limiting expression of the personalized path timeliness impact value is as follows:

[0079]

[0080] Where PUV represents the cumulative value of progress update, IUD represents the portrait update duration, and IUD0 represents the reference portrait update duration. Indicates progress update allocation value, Indicates the progress determination allocation value, It represents the portrait update allocation value, and PPT represents the personalized path timeliness impact value.

[0081] In the embodiment, the algorithm analyzes the personalized path timeliness influence value by combining the progress determination deviation amount, the progress update cumulative value, the image update time length, and the corresponding path timeliness distribution value. In the formula, when the progress update cumulative value is larger, it indicates that the timeliness of the corresponding employee training image is higher, the timeliness influence on the personalized training path is lower, and the personalized path timeliness influence value is lower; when the progress determination deviation amount is larger, it indicates that the timeliness of the image construction data of the corresponding category is higher, and the personalized path timeliness influence value is lower; when the image update time length is greater than the reference image update time length, it indicates that the past time length of the employee training image update is longer, the timeliness influence on the personalized training path is greater, and the personalized path timeliness influence value is greater. Through the analysis of the personalized path timeliness influence value, it is helpful to more accurately quantify the influence degree of the current employee training image on the personalized training path to be generated, so as to optimize the personalized training path generation interval, and further improve the timeliness of the personalized training path.

[0082] As a further scheme, the specific process of optimizing the personalized training path generation interval is as follows:

[0083] R1, the personalized path timeliness influence value and the path timeliness influence limit value are subjected to difference operation to obtain the timeliness influence difference.

[0084] R2, the path generation interval influence multiple of the next personalized training path is obtained by mapping the timeliness influence difference in the preset database.

[0085] Specifically, the timeliness mapping set between the timeliness influence difference and the path generation interval influence multiple is constructed in the preset database. The real-time timeliness influence difference is input into the timeliness mapping set to output the corresponding path generation interval influence multiple. The timeliness mapping set represents a set of mapping relationships between the timeliness influence difference and the path generation interval influence multiple.

[0086] R3, the path generation interval influence multiple and the interval time length of the last personalized training path generation are subjected to compensation operation to obtain the corresponding optimized path reconstruction interval.

[0087] Wherein, the compensation operation represents the product operation of the path generation interval influence multiple and the interval time length of the last personalized training path generation.

[0088] R4, after the optimized path reconstruction interval, the personalized training path construction is performed again.

[0089] In this embodiment, through difference calculation and mapping of timeliness impact differences, the system can dynamically identify changes in employee progress during training and timely adjust the generation interval of personalized training paths. At the same time, the system will dynamically shorten the generation interval of personalized training paths, thereby ensuring the timeliness of employee training portraits, thereby improving the timeliness of the generation of personalized training plans.

[0090] like Figure 3 As shown, it is a flowchart for determining whether to adjust the initial portrait update priority provided by an embodiment of the present application. The specific logic is: perform a reconstruction effectiveness evaluation on the reconstructed personalized training path to obtain the corresponding training path effectiveness evaluation index, and compare the training path effectiveness evaluation index with the path-portrait conformity judgment value: if the training path effectiveness evaluation index is less than the path-portrait conformity judgment value, then the employee training portrait update impact rate is obtained based on the training path effectiveness evaluation index mapping, and the employee training portrait update impact rate is compensated with the initial portrait update priority of each type of portrait construction data to obtain the frequency compensation amount of each type of portrait construction data, and the frequency compensation amount is compensated with the initial update frequency of each type of portrait construction data to obtain the optimized update frequency, and each type of portrait construction data is updated based on the optimized update frequency; if the training path effectiveness evaluation index is not less than the path-portrait conformity judgment value, the initial portrait update priority is not adjusted; through the above process, it is possible to accurately determine whether it is necessary to adjust the initial portrait update priority of the employee training portrait, thereby improving the timeliness of the employee training portrait and the timeliness of the generation of personalized training plans.

[0091] As a further solution, the specific steps for determining whether to adjust the initial profile update priority are as follows: perform a reconstruction effectiveness evaluation on the reconstructed personalized training path to obtain a corresponding training path effectiveness evaluation index, and compare the training path effectiveness evaluation index with the path-profile conformity judgment value obtained from a preset database:

[0092] In the first case, if the training path effectiveness evaluation index is less than the path-profile conformity judgment value, the employee training portrait update impact rate is obtained based on the training path effectiveness evaluation index mapping, and the employee training portrait update impact rate is compensated with the initial portrait update priority of each type of portrait construction data to obtain the frequency compensation amount of each type of portrait construction data. The frequency compensation amount is compensated with the initial update frequency of each type of portrait construction data to obtain the optimized update frequency, and each type of portrait construction data is updated based on the optimized update frequency.

[0093] It is necessary to supplement that the path mapping set between the training path effectiveness evaluation index and the employee training portrait update influence rate is constructed in the preset database, the real-time training path effectiveness evaluation index is input into the path mapping set, and the employee training portrait update influence rate is output. The path mapping set represents a set of mapping relationships between the training path effectiveness evaluation index and the employee training portrait update influence rate.

[0094] Secondly, the compensation operation represents the product operation of the employee training portrait update influence rate and the initial portrait update priority of each type of portrait construction data.

[0095] In the second case, if the training path effectiveness evaluation index is not less than the path-portrait coincidence degree determination value, the initial portrait update priority is not adjusted.

[0096] In this embodiment, by evaluating the effectiveness of the personalized training path, the system can dynamically adjust the priority of employee portrait update. When the training path effectiveness is lower, the priority of employee training portrait update can be improved by optimizing the update frequency to ensure that the employee training portrait is always associated with the training progress. According to the training path effectiveness evaluation index and the path-portrait coincidence degree determination value, the system can accurately determine whether to adjust the initial portrait update priority of the employee training portrait. Through the adjustment mechanism, the employee training portrait is not lagging behind, the timeliness of the employee training portrait is improved, and the timeliness of the personalized training scheme is also improved.

[0097] As a further scheme, the specific acquisition process of the training path effectiveness evaluation index is as follows:

[0098] A1, obtaining training path progress data, the training path progress data including average training index completion time, training progress percentage, training participation time length and training index completion rate.

[0099] Specifically, the average training index completion time is obtained by mean operation of the training index completion time recorded by the system log, the training progress percentage is obtained by ratio operation of the current training index completion task quantity and the total task quantity, the training participation time length is obtained by the system log statistics, the number of training indexes of all completed tasks is counted, and the training index completion rate is obtained by ratio operation of the total training index data.

[0100] A2, obtaining training path progress distribution rate and reference path progress time data from the preset database, the training path progress distribution rate including index completion time distribution rate, training progress distribution rate, training time length distribution rate and index completion distribution rate, the reference path progress time data including reference training index completion time and reference training participation time length.

[0101] Specifically, the training path progress allocation rate is obtained from the preset database, and the training path progress allocation rate represents the influence degree of the training path progress data on the training path effectiveness evaluation index. Each training path progress data has a unique mapping relationship with the training path progress allocation rate, and the value range is between 0 and 1; for example, a mapping set of the training path progress data and the preset training path progress allocation rate is constructed, and the real-time average training index completion time, the training progress percentage, the training participation time length and the training index completion rate are input into the mapping set to obtain the corresponding index completion time allocation rate, the training progress allocation rate, the training time length allocation rate and the index completion allocation rate, which respectively represent the influence degree of the average training index completion time, the training progress percentage, the training participation time length and the training index completion rate on the training path effectiveness evaluation index, and the sum of the four is 1.

[0102] Specifically, the reference path progress time data is preset by a preset staff and stored in the preset database.

[0103] A3, the average training index completion time is subjected to ratio operation with the reference training index completion time to obtain a training index completion time value, and the training participation time length is subjected to ratio operation with the reference training participation time length to obtain a training participation value.

[0104] It should be noted that the training index completion time value is The training participation value is

[0105] A4, the training path effectiveness evaluation index is obtained by coupling after weighting operation based on the training index completion time value, the training participation value, the training progress percentage, the training index completion rate and the corresponding training path progress allocation rate, and the training path effectiveness evaluation index is used to quantify the effective degree of the optimized and reconstructed personalized training path.

[0106] The specific limit expression of the training path effectiveness evaluation index is as follows:

[0107]

[0108] In the formula, ACT represents the average training index completion time, TPP represents the training progress percentage, TID represents the training participation time length, CTI represents the training index completion rate, ACT0 represents the reference training index completion time, TID0 represents the reference training participation time length, represents the index completion time allocation rate, represents the training progress allocation rate, represents the training time length allocation rate, represents the index completion allocation rate, and TPE represents the training path effectiveness evaluation index.

[0109] In the embodiment, the algorithm analyzes the training path progress data, the training path progress distribution rate and the reference path progress time data to obtain a training path effectiveness evaluation index. As the training index completion time value, the training participation value, the training progress percentage and the training index completion rate increase, the training progress of the employee based on the current personalized training path learning is faster, the training quality is higher, and the training path effectiveness evaluation index is larger. Through the analysis of the training path effectiveness evaluation index, the effectiveness of the optimized and reconstructed personalized training path is more accurately quantified, the generation timeliness of the personalized training scheme and the sufficiency of the training progress correlation degree are improved, and the individuality of the personalized training scheme and the accuracy of the corresponding training scheme of the employee are ensured.

[0110] It should be noted that the training path effectiveness evaluation index can evaluate the effectiveness of the generated reconstructed personalized training path in a timely manner, timely feedback the training effect of the personalized path, and timely optimize the initial portrait update priority, so that the generation of the personalized training scheme forms a closed loop, thereby improving the correlation between the personalized training path and the employee training portrait and improving the timeliness of the personalized training scheme.

[0111] As shown in Figure 4 The flowchart of the personalized training scheme generation method based on data analysis provided by the embodiment of the application is shown in FIG. 1. The specific logic is as follows: first, obtain the initial portrait update priority of each type of portrait construction data from a preset database, and evaluate the training progress change of the employee training portrait based on the change of each type of portrait construction data; then, determine whether to trigger portrait update based on the obtained evaluation result, if yes, update the portrait construction data of the corresponding category, otherwise, perform update critical assessment on each type of portrait construction data; then, determine whether to reconstruct the employee training portrait based on the obtained evaluation result, construct a personalized training path based on the obtained employee training portrait, and perform update timeliness quantification on the employee training portrait; finally, determine whether to reconstruct the personalized training path based on the quantification result, if not, optimize the personalized training path generation interval, otherwise, perform reconstruction effectiveness evaluation on the reconstructed personalized training path to determine whether to adjust the initial portrait update priority; through the above process, the timeliness of the employee training portrait is improved, and the generation timeliness of the personalized training scheme is also improved.

[0112] The embodiment of the application also provides a personalized training scheme generation method based on data analysis, and the specific steps are as follows:

[0113] The initial image update priority of various types of image construction data is obtained from a preset database, and the training progress change of the employee training image is evaluated in combination with the change of the various types of image construction data. The employee training image is used to measure the progress change of the employee in the actual training process.

[0114] Based on the obtained evaluation result, it is judged whether to trigger image update. If yes, the image construction data of the corresponding category is updated, otherwise the update threshold of various image construction data is evaluated.

[0115] Based on the obtained evaluation result, it is judged whether to reconstruct the employee training image. Based on the obtained employee training image, a personalized training path is constructed, and the timeliness of the employee training image is quantified.

[0116] Based on the quantification result, it is judged whether to reconstruct the personalized training path. If not, the interval of generating the personalized training path is optimized, otherwise the reconstruction effectiveness of the reconstructed personalized training path is evaluated to determine whether to adjust the initial image update priority.

[0117] In this embodiment, the specific embodiment of data analysis in the personalized training scheme is to collect, mine and analyze employee data to design a high-correlation personalized training path for each employee, and continuously track and evaluate the training effect. The personalized training scheme generation method based on data analysis combines accurate image update, personalized path optimization, timeliness quantification and reconstruction effectiveness evaluation, ensuring that the training content is closely combined with the actual training progress of the employee, and avoiding outdated or ineffective training content. At the same time, this method not only improves the timeliness of the employee training image, but also improves the timeliness of the generation of the personalized training scheme.

[0118] It should be noted that in the personalized training scheme based on data analysis, the process of constructing the employee training image is very critical, which helps to customize the training plan for each employee individually, which is mainly realized through the following main steps: first, collect data, including but not limited to personal information, work performance data, training history and behavior data; then feature extraction and standardization are performed on the collected data; then analysis is performed through data modeling, common techniques include clustering analysis, regression analysis, association rule analysis and deep learning and artificial intelligence; finally, image construction is performed based on data modeling, the employee training image usually includes basic information dimension, ability dimension, development demand dimension and behavior characteristic dimension, etc.; In addition, based on the analysis result of the employee training image, a personalized training scheme is constructed and recommended to the client of the corresponding employee, and the timeliness of the employee training image is evaluated based on the progress feedback of the employee training scheme, therefore, the construction of the employee training image and the personalized training path influences each other, through the timeliness analysis of the employee training image, it is helpful to improve the timeliness of the personalized training path of the employee.

[0119] It needs to be pointed out that, as Figure 5 shown, the interface diagram of the personalized training scheme generation system based on data analysis provided by the embodiment of the application is shown, the left side of the interface is the system navigation bar, containing the options of "home page", "training data analysis", "training plan management", "execution and feedback", "report center" and "system setting". The upper middle part is the group style display area, which updates the group dynamics in real time. The lower middle part is recommended for you, which recommends training tasks of different difficulty levels and other information to the user. It is convenient for the user to quickly enter the training and improve the training progress.

[0120] As Figure 6 shown, the interface diagram of the training portrait initial evaluation module of the personalized training scheme generation system based on data analysis provided by the embodiment of the application is shown, the left side of the interface is the navigation bar of the training portrait initial evaluation module. The left area is the state display area of "initial portrait update priority", which presents a variety of data types and corresponding initial portrait update priorities in the form of a list. The right area is the analysis result area, which is displayed in the form of a curve graph, showing the progress change trend of each training index, so that the user can more intuitively view the progress of each training index.

[0121] As Figure 7 shown, the interface diagram of the portrait reconstruction determination module of the personalized training scheme generation system based on data analysis provided by the embodiment of the application is shown, the left side of the interface is the portrait reconstruction determination module navigation bar, the middle left side is the speed change difference column chart of various portrait construction data. The right side of the middle is the corresponding progress update value part, wherein the "first progress update value" corresponds to the "data pre-update score", the "second progress update value" corresponds to the "data waiting update score", and the "progress update cumulative value" corresponds to the "portrait update score", so that the user can more intuitively understand the update situation of the current employee training portrait.

[0122] As Figure 8 shown, the interface diagram of the training path timeliness determination module of the personalized training scheme generation system based on data analysis provided by the embodiment of the application is shown, the left side of the interface is the training path timeliness determination module navigation bar. The left side of the middle is the change curve graph of the individualized path timeliness influence value, and the right side of the middle is the progress of the reconstruction of the individualized training path, which is updated in real time, and is helpful to more intuitively understand the progress of the update of the individualized training path.

[0123] As Figure 9As shown, it is the interface diagram of the reconstruction path effectiveness evaluation module of the personalized training scheme generation system based on data analysis provided by the embodiment of the present application, the left side of the interface is the navigation bar of the reconstruction path effectiveness evaluation module. The left side of the middle is the related data display of the initial image update priority optimization, and the right side of the middle is the curve diagram of the training path effectiveness evaluation index, which is convenient for the user to understand whether the current personalized training path is more time-effective.

[0124] To sum up, the embodiment of the present application combines the change of various types of image construction data to evaluate the training progress change of the employee training image by the obtained initial image update priority, and then judges whether to trigger image update. If yes, update, otherwise, perform update critical assessment, then judge whether to reconstruct the employee training image based on the evaluation result, then reconstruct the personalized training path, and update the time effectiveness of the employee training image, and then judge whether to reconstruct the personalized training path, if not, optimize the personalized training path generation interval, otherwise, judge whether to adjust the initial image update priority, thereby improving the timeliness of the employee training image, and further improving the timeliness of the personalized training scheme generation, effectively solving the problem of insufficient correlation between the timeliness of the personalized training scheme generation and the training progress in the prior art.

[0125] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

Claims

1. A personalized training program generation system based on data analysis, characterized by: It includes the training portrait initial evaluation module, the portrait reconstruction determination module, the training path timeliness determination module and the reconstruction path effectiveness evaluation module: The training portrait initial assessment module is used to obtain the initial portrait update priority of various portrait construction data from a preset database, and evaluate the training progress changes of employee training portraits based on the changes in various portrait construction data. The employee training portrait is used to measure the progress changes of employees in the actual training process; The portrait reconstruction determination module is used to determine whether to trigger a portrait update based on the obtained evaluation results. If so, the portrait construction data of the corresponding category is updated; otherwise, a critical assessment is performed on the update of each category of portrait construction data; The training path timeliness determination module is used to determine whether to rebuild the employee training profile based on the obtained evaluation results, build a personalized training path based on the obtained employee training profile, and quantify the update timeliness of the employee training profile; The reconstruction path effectiveness evaluation module is used to determine whether to reconstruct the personalized training path based on the quantitative results. If not, the personalized training path generation interval is optimized. Otherwise, the reconstruction effectiveness evaluation of the reconstructed personalized training path is performed to determine whether the initial portrait update priority is adjusted.

2. The system for generating personalized training programs based on data analysis according to claim 1, characterized in that: The specific steps for evaluating changes in training progress for employee training profiles are as follows: Classify employee profile construction data and obtain the historical average update frequency corresponding to each type of profile construction data. Based on the historical average update frequency, map the initial profile update priority of the corresponding category of profile construction data in the preset database; Real-time monitoring of changes in various profile building data to obtain the corresponding impact of changes in employee training progress; Determine the impact of employee training progress changes and the progress change critical value obtained from the preset database: If the impact of the change in employee training progress is not less than the critical value of progress change, the portrait construction data of the corresponding category will be updated; otherwise, the changes in the portrait construction data of the corresponding category will continue to be monitored.

3. The system for generating personalized training programs based on data analysis according to claim 2, characterized in that: The specific method for obtaining the impact of employee training progress change is as follows: Obtaining portrait construction data change parameters, wherein the portrait construction data change parameters include initial portrait update priority and training indicator change quantity; Obtaining a portrait construction change allocation from a preset database, wherein the portrait construction change allocation includes an initial priority allocation and a training indicator allocation; After weighted operation on the portrait construction data change parameters and the portrait construction change allocation amount, the employee training progress change impact amount is obtained. The employee training progress change impact amount is used to quantify the impact of employee training indicators on employee training portraits.

4. The system for generating personalized training programs based on data analysis according to claim 1, characterized in that: The specific process of updating the critical assessment of various types of portrait construction data is as follows: Obtaining progress change differences of various types of profile building data that have not been updated, and comparing each progress change difference with a preset critical value obtained from a preset database, wherein the progress change difference represents the difference between the employee training progress change impact amount and the progress change critical value; If the progress change difference of the portrait construction data of the corresponding category is lower than the preset critical value, the progress update value of the portrait construction data of the corresponding category is recorded as the first progress update value; otherwise, the progress update value of the portrait construction data of the corresponding category is recorded as the second progress update value; The progress update values ​​of all categories of portrait construction data are counted to obtain the progress update cumulative value. If the progress update cumulative value is higher than the preset cumulative value, the employee training portrait is updated, otherwise no additional processing is performed.

5. The system for generating personalized training programs based on data analysis according to claim 1, characterized in that: The specific process for quantifying the timeliness of updating employee training profiles is as follows: When generating a personalized training path, obtain the current update status of various profile construction data to obtain the personalized path timeliness impact value of the personalized training path, and compare the current personalized path timeliness impact value with the path timeliness impact limit value obtained from the preset database: If the personalized path timeliness impact value is not greater than the path timeliness impact limit, the personalized training path generation interval is optimized; If the personalized path timeliness impact value is greater than the path timeliness impact limit, the reconstruction effectiveness evaluation of the reconstructed personalized training path will be performed.

6. The system for generating personalized training programs based on data analysis according to claim 5, characterized in that: The specific process of obtaining the personalized path timeliness impact value is as follows: Obtaining path timeliness evaluation data, including the cumulative value of progress updates, the impact of employee training progress changes, the critical value of progress changes, and the duration of the profile update; Obtaining a path timeliness distribution value and a reference portrait update duration from a preset database, wherein the path timeliness distribution value includes a progress update distribution value, a progress determination distribution value, and a portrait update distribution value; Perform deviation calculation on the employee training progress change impact amount and the progress change critical value to obtain the progress judgment deviation amount; The progress judgment deviation is weighted and coupled with the progress update cumulative value, the portrait update duration, and the corresponding path timeliness allocation value to obtain the personalized path timeliness impact value. The personalized path timeliness impact value is used to quantify the degree of influence of the current employee training profile on the personalized training path to be generated.

7. The system for generating personalized training programs based on data analysis according to claim 6, characterized in that: The specific process of optimizing the personalized training path generation interval is as follows: Perform a difference operation on the personalized path timeliness impact value and the path timeliness impact limit value to obtain the timeliness impact difference; The path generation interval impact multiple of the next personalized training path is obtained by mapping the timeliness impact difference in the preset database; Compensate the path generation interval impact multiple with the interval duration of the last one-time training path generation to obtain the corresponding optimized path reconstruction interval; After optimizing the path reconstruction interval, personalized training path construction is performed again.

8. The system for generating personalized training programs based on data analysis according to claim 1, characterized in that: The specific steps of determining whether to adjust the initial portrait update priority are as follows: The reconstruction effectiveness evaluation of the reconstructed personalized training path is performed to obtain the corresponding training path effectiveness evaluation index, and the training path effectiveness evaluation index is compared with the path-profile conformity judgment value obtained from the preset database: If the training path effectiveness evaluation index is less than the path-profile conformity judgment value, the employee training profile update impact rate is obtained based on the training path effectiveness evaluation index mapping, and the employee training profile update impact rate is compensated with the initial profile update priority of each type of profile construction data to obtain the frequency compensation amount of each type of profile construction data. The frequency compensation amount is compensated with the initial update frequency of each type of profile construction data to obtain the optimized update frequency, and the various types of profile construction data are updated based on the optimized update frequency; If the training path effectiveness evaluation index is not less than the path-profile conformity judgment value, the initial profile update priority will not be adjusted.

9. The system for generating personalized training programs based on data analysis according to claim 8, characterized in that: The specific process of obtaining the training path effectiveness evaluation index is as follows: Obtaining training path progress data, including average training indicator completion time, training progress percentage, training participation time, and training indicator completion rate; Obtaining training path progress allocation rate and reference path progress time data from a preset database, wherein the training path progress allocation rate includes the indicator completion time allocation rate, the training progress allocation rate, the training duration allocation rate, and the indicator completion allocation rate; and the reference path progress time data includes the reference training indicator completion time and the reference training participation duration; The training indicator completion time value is obtained by performing a ratio operation on the average training indicator completion time and the reference training indicator completion time, and the training participation time is obtained by performing a ratio operation on the reference training participation time; The training path effectiveness evaluation index is obtained by coupling the training indicator completion time value, training participation value, training progress percentage, training indicator completion rate, and the corresponding training path progress distribution rate after weighted calculation. The training path effectiveness evaluation index is used to quantify the effectiveness of the personalized training path after optimization and reconstruction.

10. A method for generating a personalized training program based on data analysis, characterized by: The specific steps are as follows; Obtaining the initial profile update priority of various profile construction data from a preset database, and evaluating employee training profiles for changes in training progress based on changes in various profile construction data. The employee training profiles are used to measure employee progress changes during the actual training process. Based on the obtained evaluation results, it is determined whether to trigger the portrait update. If so, the portrait construction data of the corresponding category is updated. Otherwise, the update critical assessment of each type of portrait construction data is performed; Based on the evaluation results, determine whether to rebuild the employee training profile, build a personalized training path based on the obtained employee training profile, and quantify the update time of the employee training profile; Based on the quantitative results, it is determined whether to reconstruct the personalized training path. If not, the personalized training path generation interval is optimized. Otherwise, the reconstruction effectiveness of the reconstructed personalized training path is evaluated to determine whether the initial portrait update priority is adjusted.

Images